From: stephanie snedden
Subject: Re: 2.5m obslog 20070424 (MJD 54215)
Date: Wed, 25 Apr 2007 17:37:31 -0600
Dan,
Sorry, Dan, I'm afraid I don't think I should be publicly rebuked,
even mildly.  I was making safe cartridge changes and doing  very
little to endanger the cartridge, spectrographs or telescope, in
exactly the same manner as I have for years. I don't think the need
to do shakedown tests (however short) outweighs taking efficient
science data in this case, since we have been using the manual mode
for years with only one incident that caused no damage. I don't mean
to belittle the need for the interlocked lift, I just don't see why
all of a sudden it has become so critical to shake it out during
science time.
It is obvious to me after seeing the interlocked lift failures over
the last nights that the system is indeed still in shakedown. I am
not willing to start authorizing shake tests, however short, on
nights when we should be doing science. Our policy is to roll back to
the system that works until new tests can be properly scheduled for
the proper shakedown  time. Let's not go down this path. However, I
won't rebuke you even mildly about this because I understand full
well how little time for testing we have left in our schedule.
Let's talk about scheduling some tests for a cloudy night, if there
is no time during the day.
Steph
On Apr 25, 2007, at 9:12 AM, Dan Long wrote:
>>
>>
>> Cartridge 2 wouldn't latch with the interlocked lift pendant.
>> Soory, Dan.
>>
>> ---=== OBSERVING LOG ===---
>>
>>
>>
>> Dan, cartridge 5 unmounted without problems, but the latches would
>> not fire
>> for mounting cartridge 2. Sorry. I had no time to deal with the
>> interlocks
>> tree and went to manual mode.
>>
>
> Hi Steph,
>
>    I feel a mild rebuke is in order regarding your comment about no
> time to check the interlocks.  Believe me, I understand the
> pressure to get going and take data, but this needs to be balanced
> with the safety of the telescope and identifying (especially
> intermittent) problems so things can be corrected.  A check of the
> interlocks takes less than a minute, and once the problem is
> identified an informed decision can be made on how to proceed.  If
> the interlocks state is noted in the log (or over the phone when
> real time debugging is needed) it has the added bonus of helping
> the day staff understand and correct the problem.
>
>    We really need to get everyone in the habit of checking the
> interlocks when a problem occurs (and trained in the navigation of
> the software so it can be done quickly).  In this case, it almost
> certainly was the instrument ID switch and once you knew that,
> switching to manual was the appropriate action.  In many cases
> though, there isn't the choice of switching to manual, and in no
> case is it a good idea to bypass an interlock if you don't know
> what is wrong.
>
>    I apologize about cartridge 2.  It looks like maybe something
> happened to the switch on the telescope rather than the mating part
> on cartridge 5.  I will check the switch and try all cartridges on
> the telescope to make sure their IDs register properly (which I
> should have done yesterday).
>
>
> 				Cheers,
>
> 				  Dan